Gemini Robotics ER 2 ranks first on 3 of the 3 benchmarks with five or more models scored.
Google DeepMind Released July 30th, 2026 Family Gemini Robotics The lab's model card Compare with others Google DeepMind Jul 2026 Closed weights Mid-size
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ERQA | 78.5% | #1 of 36 | Lab model card | |
| Instrument reading (Google DeepMind) | 65.7% | #1 of 5 | Official leaderboard | |
| Progress classification (Google DeepMind) | 57.4% | #1 of 5 | Official leaderboard | |
| ASIMOV | 97.9% | #1 of 2 | Lab model card | |
| Success detection (Google DeepMind) | 87.7% | #2 of 2 | Lab model card | |
| Human proximity (Google DeepMind robotics) | 93% | #1 of 1 | Lab model card | |
| Moment finding (Google DeepMind) | 91.3% | #1 of 1 | Lab model card |
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
| Benchmark | Lab says | Board says | Gap |
|---|---|---|---|
| Human proximity (Google DeepMind robotics) | 93% Human proximity (1m) | 93% Leaderboard | Same |
| Instrument reading (Google DeepMind) | 65.7% Generalized instrument reading | 65.7% Leaderboard | Same |
| Moment finding (Google DeepMind) | 91.3% Moment-finding accuracy (0.96s mean absolute distance) | 91.3% Leaderboard | Same |
| Progress classification (Google DeepMind) | 57.4% Progress classification (five bands) | 57.4% Leaderboard | Same |
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 1 of them isn't in the launch post.
